Linux

Troubleshooting Cases for No Redundant Path of Huawei-Developed UltraPath

  1. On the CLI of the application server, run the upadm show version command to check whether the version after UltraPath for Linux is 5.01.017 or earlier.

    • If yes, go to step 2.
    • If no, go to step 8.

  2. Check the controller enclosure type by referring to section "Connecting Cables" in the OceanStor 5300 V3&5500 V3&5600 V3&5800 V3&6800 V3 Storage System V300R003 Installation Guide. (You can download this document from Enterprise or Carrier. V300R003 is just used as an example, you can download the Installation Guide of the corresponding version from Huawei technical support website.) Then, check whether the controller enclosure is connected to application servers using redundant paths by referring to section "Connecting the Controller Enclosure to Application Servers."

    • If yes, go to step 3.
    • If no, go to step 4.

  3. On the CLI of the application server, run the upadm show path command and record the command output. Then, connect cables by referring to section "Connecting the Controller Enclosure to Application Servers" in the OceanStor 5300 V3&5500 V3&5600 V3&5800 V3&6800 V3 Storage System V300R003 Installation Guide. (You can download this document from Enterprise or Carrier.)

    • If yes, go to step 4.
    • If no, go to step 18.

  4. For all disks having no redundant path, run the upadm show lun array=xx dev=yy command (replace xx with a value obtained in the Array ID column of the command output after upadm show array is executed, replace yy with the ID of a disk having no redundant path) on the CLI of the application server to check whether all logical paths of the disk have been FAILED.

    • If yes, go to step 5.
    • If no, go to step 6

  5. The disk has been FAILED on the application server, consult the user whether the disk has been used.

    • If yes, go to step 18.
    • If no, ignore current disk and perform Step 4 again to check next disk having no redundant util all disks are checked.

  6. For all disks having no redundant path, consult the user whether logical paths have been manually deleted.

    • If yes, go to step 7.
    • If no, no further action is required.

  7. On the CLI of the application server, run the upadm show lun array=xx command (replace xxx with a value obtained in the Array ID column of the command output after upadm show array is executed) and record the command output. Then, run the hot_add command to scan for disks during off-peak hours. After the command is executed, run the upadm show lun array=xx command again to check whether a logical path whose Path State is OPTIMAL.

    • If yes, no further action is required.
    • If no, go to step 18.

  8. On the CLI of the application server, run the upadmin show path command and record the command output. Then, connect cables by referring to section "Connecting the Controller Enclosure to Application Servers" in the OceanStor 5300 V3&5500 V3&5600 V3&5800 V3&6800 V3 Storage System V300R003 Installation Guide. (You can download this document from Enterprise or Carrier. V300R003 is just used as an example, you can download the Installation Guide of the corresponding version from Huawei technical support website.) For HyperMetro system, also connect cables by referring to section "Installation -> Connecting Cables" in the OceanStor 5300 V3&5500 V3&5600 V3&5800 V3&6800 V3 Storage System V300R003 HyperMetro Feature Guide. (You can download the document from Enterprise or Carrier. V300R003 is just used as an example, you can download the Hypermetro Feature Guide of the corresponding version from Huawei technical support website.) Then, run the hot_add command to scan for disks during off-peak hours. After the command is executed, run the upadmin show path command again to check whether a logical path whose Path State is Normal is displayed.

    • If yes, go to step 9.
    • If no, go to step 18.

  9. For all disks having no redundant path, run the upadmin show vlun id=xx command (replace xx with the ID of a disk having no redundant path, run the upadm show vlun id=xx type=all command for a HyperMetro LUN instead) on the CLI of the application server to check whether all logical paths of the disk have been Fault（check whether two aggregation members of a HyperMetro LUN have been Fault）.

    • If yes, go to step 10.
    • If no, go to step 11.

  10. The disk has been Fault on the application server, consult the user whether the disk has been used.

    • If yes, go to step 18.
    • If no, ignore current disk and perform Step 9 again to check next disk having no redundant util all disks are checked.

  11. For all disks having no redundant path, consult the user whether logical paths have been manually deleted.

    • If yes, go to step 12.
    • If no, go to step 13.

  12. On the CLI of the application server, run the upadmin show vlun id=xx command (replace xxx with a value obtained in the check result, run the upadmin show vlun id=xx type=all command for a HyperMetro system) and record the command output. Then, run the hot_add command to scan for disks during off-peak hours. After the command is executed, run the upadmin show vlun id=xx (run the upadmin show vlun id=xx type=all command for a HyperMetro system) command again to check whether a logical path whose Path State is Normal.

    • If yes, go to step 13.
    • If no, go to step 18.

  13. On the CLI of the application server, run the upadmin show array id=xx command (replace xx with the value obtained in the Array ID column of the command output after upadmin show array is executed) to check whether Status of a controller is Disable.

    • If yes, go to step 14.
    • If no, go to step 15.

  14. Consult the user whether the disabled controllers can be enabled. If the user allows you to do so, on the CLI of the application server, run the upadmin set tpgstate=enable array_id=xx tpg_id=yy command (for example, xx is 0 and yy is 0A) to enable all controllers and check whether the controller states are Enable and path states of controllers are Normal.

    • If yes, go to step 15.
    • If no, go to step 18.

  15. On the CLI of the application server, run the upadmin show path command to check whether the physical path state is Disable.

    • If yes, go to step 16.
    • If no, go to step 17.

  16. Consult the user whether the disabled paths can be enabled. If the user allows you to do so, on the CLI of the application server, run the upadmin set pathstate=enable path_id=xx command (the value of path_id is the same as that of Path ID in step 10) to enable all physical paths and check whether the physical path states are Normal.

    • If yes, go to step 17.
    • If no, go to step 18.

  17. On the CLI of the application server, run the upadmin show path command to check whether the physical path state in the Path State column contains key word degradation.

    • If yes, go to step 18.
    • If no, no further action is required.

  18. Collect related event information using InfoGrab, and send it to technical support engineers for further handling.

Troubleshooting Cases for No Redundant Path of MultiPath Built in Linux

  1. Check the controller enclosure type by referring to section "Connecting Cables" in the OceanStor 5300 V3&5500 V3&5600 V3&5800 V3&6800 V3 Storage System V300R003 Installation Guide. (You can download this document from Enterprise or Carrier. V300R003 is just used as an example, you can download the Installation Guide of the corresponding version from Huawei technical support website.) Then, check whether the controller enclosure is connected to application servers using redundant paths by referring to section "Connecting the Controller Enclosure to Application Servers."

    • If yes, go to step 3.
    • If no, go to step 2.

  2. On the CLI of the application server, run the multipath -ll command and record the command output. Then, connect cables by referring to section "Connecting the Controller Enclosure to Application Servers" in the OceanStor 5300 V3&5500 V3&5600 V3&5800 V3&6800 V3 Storage System V300R003 Installation Guide. (You can download this document from Enterprise or Carrier.) And run the multipath -v2 command to scan for disks during off-peak hours. After the command is executed, run the multipath -ll command again to check whether a logical path whose Path State is active or enabled is displayed, and the next line contains ready running.

    • If yes, go to step 3.
    • If no, go to step 5.

  3. For all disks having no redundant path, consult the user whether logical paths have been manually deleted.

    • If yes, go to step 4.
    • If no, no further action is required.

  4. On the CLI of the application server, run the multipath -ll command and record the command output. Then, run the multipath -v2 command to scan for disks during off-peak hours. After the command is executed, run the multipath -ll command again to check whether a logical path whose Path State is active or enabled is displayed, and the next line contains ready running.

    • If yes, no further action is required.
    • If no, go to step 5.

  5. Collect related event information using InfoGrab, and send it to technical support engineers for further handling.

Huawei Proprietary and Confidential
Copyright © Huawei Technologies Co., Ltd.